The Mitochondrial Protein RESISTANCE to APHIDS 9 Interacts with S40 to Resist Aphid Infestation by Modulating Reactive Oxygen Species Homeostasis in Maize (Zea mays)

Abstract

As the corn aphid (Rhopalosiphum maidis) poses a major threat to maize (Zea mays) growth, there is much interest in identifying aphid resistance genes. In this study, an aphid‐susceptible maize mutant from an ethyl methanesulfonate–mutagenized library is identified that exhibits greater aphid settlement than the wild type. Using the MutMap approach, the causal gene RESISTANCE TO APHIDS 9 (RTA9) is cloned, which encodes a mitochondrion‐localized protein from the Domain of Unknown Function 641 family. Overexpressing RTA9 in maize confers significant resistance to aphids without compromising seed yield. It further identifies the senescence regulator S40 as an interactor of RTA9, which negatively regulates the stability of S40. Knockout of S40 enhanced aphid resistance, while its overexpression increased susceptibility. Further analysis demonstrates that the rta9‐1 mutant does not exhibit significant enrichment of differentially expressed genes associated with oxidoreductase activity following aphid infestation. By contrast, genes involved in this pathway are significantly enriched in the s40 mutant. Additionally, aphid‐induced reactive oxygen species (ROS) levels are markedly lower in rta9‐1 than in the wild type but significantly higher in s40. Collectively, the results suggest that the mitochondrial protein RTA9 and its interacting partner S40 regulate resistance to aphid infestation by modulating ROS homeostasis.

This study identifies the mitochondrial protein RTA9, a member of the Domain of Unknown Function 641 family, together with its partner S40, as key regulators of aphid resistance in maize. RTA9 promotes S40 degradation and modulates reactive oxygen species accumulation. Either overexpression of RTA9 or knockout of S40 enhances aphid resistance without yield penalties, offering promising genetic targets for developing aphid‐resistant crops.

1. Introduction

Aphids (Hemiptera: Aphididae) are prominent and destructive pests in crop production that obtain nutrients from their host by piercing the plant tissues and sucking the phloem sap.^{[ 1 ]} Due to their rapid reproduction cycle, aphid infestations can spread quickly, severely compromising crop nutrient acquisition and suppressing plant growth.^{[ 2 ]} Aphids also secrete honeydew, which can adhere to the surface of leaves and, in maize (Zea mays), to the tassels, resulting in lower photosynthetic rate and pollination efficiency, respectively, and negatively affecting crop yield and quality.^{[ 3} , ^{4 ]} Aphids are also vectors for certain viruses and can transmit diseases that reduce yield or even lead to plant death.^{[ 5} , ⁶ , ^{7 ]}

Maize is an important cereal crop worldwide, serving as a staple source of human food, animal feed, and industrial materials.^{[ 8} , ^{9 ]} The corn aphid Rhopalosiphum maidis has become a major cause of yield loss in maize production.^{[ 10 ]} In agriculture, breeding aphid‐resistant varieties is the most ecologically‐friendly and effective approach to alleviating the threats posed by this pest.^{[ 11 ]} Extensive efforts have been devoted to dissecting the genetic basis and molecular mechanisms of aphid resistance in maize, yet due to the complexity of aphid populations, their diverse hosts and the influences of local environmental factors in the field, progress in aphid control research has been limited.

Plants are constantly evolving physical and chemical barriers that prevent insects from successfully colonizing and feeding on their tissues.^{[ 1} , ¹² , ^{13 ]} Epicuticular waxes and glandular trichomes provide a physical barrier against non‐adapted aphids,^{[ 14 ]} while several plant chemical compounds have anti‐nutritive, repellent, or toxic effects on aphids, such as callose,^{[ 15 ]} betulin,^{[ 16 ]} flavonoids, and free gossypol.^{[ 17 ]} Such plant substances have been the focus of research aimed at controlling aphids through multiple approaches.^{[ 18} , ^{19 ]} For instance, the ectopic expression of an agglutinin gene from the traditional Chinese herb crow‐dipper (Pinellia ternata) in tobacco (Nicotiana tabacum) conferred broad‐spectrum resistance to aphids and other insects.^{[ 19 ]}

At the molecular level, the signaling mechanisms of anti‐insect pathways in plants are important for both local immune responses and the propagation to systemic tissues, including the propagation of electrical signals, calcium influx, reactive oxygen species (ROS) production, and the plant hormones, such as salicylic acid (SA) and jasmonic acid (JA).^{[ 20 ]} Among these pathways and molecules, ROS act as a shield against insect invasion and serve as a central regulator of various biological programs, allowing cells to respond rapidly to different stimuli.^{[ 21 ]} Elevated cellular ROS levels, known as a ROS burst, enhance the plant's resistance by activating the expression of defense‐related genes,^{[ 22} , ^{23 ]} reprogramming signaling networks, and modifying plant secondary metabolic pathways to reduce insect feeding, growth, and survival.^{[ 21} , ^{24 ]} Conversely, the inhibition of ROS bursts promotes aphid colonization in the host plant.^{[ 22} , ^{25 ]} For instance, lower ascorbate oxidase activity results in a more pronounced reduction of the apoplastic redox state, compromising host resistance to aphids. This resistance manifests through multifaceted changes, including altered cell wall composition, transcriptional reprogramming of phytohormone‐mediated defense genes, and metabolic shifts in amino acids and sugars.^{[ 26 ]} In Arabidopsis (Arabidopsis thaliana), plants harboring a null mutation for RBOHD, which encodes the apoplastic NADPH oxidase RESPIRATORY BURST OXIDASE HOMOLOG D, support accelerated colony expansion of the green peach aphid (Myzus persicae) compared to their wild‐type counterparts, indicating that ROS generation is essential for mounting fully effective defense responses.^{[ 27 ]} In wheat (Triticum aestivum), H₂O₂ accumulation inhibits the proliferation of apterous (wingless) aphids while promoting the production of alate (winged) aphids.^{[ 28 ]} Through evolutionary adaptation, aphids have developed countermeasures against host defenses: the salivary effector proteins Mp55 from M. persicae and Histidine‐rich calcium‐binding protein (ApHRC) from the pea aphid (Acyrthosiphon pisum) inhibit host ROS production, significantly enhancing aphid fecundity.^{[ 25} , ^{29 ]} The salivary protein Cathepsin B3 (CathB3), a cysteine protease, suppresses feeding by triggering phloem‐localized ROS accumulation; thus, aphids that express high levels of CathB3 exhibit lower colonization success due to feeding‐induced ROS bursts, while aphids with low CathB3 expression have enhanced colonization fitness.^{[ 22 ]}

Mitochondria are the main organelles responsible for utilizing and converting chemical energy into a usable form, which requires the activity of the many core subunits of the electron transport chain (ETC). Mitochondria are a major source of ROS in plant cells.^{[ 30 ]} Recent studies have shown that several mitochondrion‐localized proteins participate in aspects of plant biology beyond energy production or conversion, such as thermotolerance, kernel development, leaf senescence, and disease resistance.^{[ 31} , ³² , ³³ , ³⁴ , ^{35 ]} Those mitochondrion‐localized proteins involved in plant tolerance or resistance are typically associated with ROS pathways; for example, the mitochondrial RNA processing factor RESISTANCE TO PHYTOPHTHORA PARASITICA 7 (RTP7) mediates plant immunity to a broad spectrum of pathogens by modulating mitochondrial ROS accumulation.^{[ 34 ]} The roles and mechanisms of mitochondrial proteins in aphid resistance are unclear in crops; however, understanding how these proteins function in this context is essential for devising strategies to enhance aphid resistance during maize breeding.

Domain of Unknown Function 641 (DUF641) is a domain found in members of a plant‐specific protein family, but its functions are unclear. While current evidence implicates DUF641‐containing proteins in regulating organ size,^{[ 36 ]} light‐dependent hypocotyl elongation,^{[ 37 ]} and drought tolerance or water usage,^{[ 38} , ^{39 ]} no association has been documented with aphid resistance to date. Here, we identified an aphid‐susceptible maize mutant and cloned the causal gene, RESISTANCE TO APHIDS 9 (RTA9; Zm00001d039444), which encodes a mitochondrion‐localized protein containing a DUF641 domain. Loss of RTA9 function impairs the ROS signaling response to aphid stimulation in maize, leading to imbalanced ROS levels. Overexpression of RTA9 conferred greater aphid resistance without negatively affecting grain yield. Further investigations revealed that RTA9 interacts with S40, a factor that regulates senescence, and affects its protein stability. Maize plants that overexpressed S40 had reduced ROS levels and attenuated aphid resistance, whereas the s40 mutant exhibited the opposite phenotype. These observations suggest that the maize mitochondrial protein RTA9 responds to aphid infestation by mediating ROS production signals, making it a potential candidate gene for developing new strategies to manage aphids in maize.

2. Results

2.1. The Maize Mutant rta9‐1 is Susceptible to Aphids

We screened an ethyl methanesulfonate (EMS)‐induced mutant maize population for aphid resistance and identified an aphid‐susceptible plant, which we designated resistance to aphids 9‐1 (rta9‐1). Many aphids colonized the sheaths and leaves of mature rta9‐1 seedlings; the susceptible plants had ≈ 10 times more aphids than the wild type, the maize inbred line B73 (Figure S1, Supporting Information).

To rule out the possibility that this phenotype might be caused by chance or environmental bias, we self‐crossed rta9‐1 and repeated the aphid bioassays on the progeny of rta9‐1 under controlled greenhouse conditions. As observed for the parental seedlings grown in the field, many aphids colonized the sheaths of rta9‐1 progeny after 21 days of infestation when grown under control conditions (Figure 1A,B). In addition, significantly more honeydew (purple‐red area on the filter paper; Figure 1C,D) was secreted by aphids feeding on rta9‐1 progeny plants than by those feeding on B73. After one week of infestation, the aphid survival rates were significantly higher on the rta9‐1 progeny seedlings than on B73 (Figure 1E). Furthermore, the aphids feeding on the rta9‐1 progeny weighed more than those feeding on the wild type (Figure 1F). Therefore, rta9‐1 appears to be an aphid‐susceptible mutant that is more suitable for aphid colonization.

Figure 1.

The rta9‐1 mutant is susceptible to aphids. A) Colonization and distribution of aphids on the wild‐type B73 and the rta9‐1 mutant after 21 days of infestation. Scale bar, 1 cm. B) Quantification of aphids growing on B73 and rta9‐1 as shown in (A). C,D) Honeydew secretion assay of aphid feeding. The quantification of the honeydew areas on the filter papers in (C) is shown in (D). E) Survival rates of aphids feeding on B73 or rta9‐1 (^** p < 0.01, ^*** p < 0.001). F) Body weight of each aphid feeding on B73 or rta9‐1. The boxes represent the interquartile range, with the middle line defining the median. The lines extending from the quartiles of the box are called “whiskers” and show the maximum and minimum values. In panels (B, D–F), at least six independent replicates were performed for each assay, and the data in panels (B, D–E) are presented as means ± standard deviation (SD). The values in (D) were normalized to B73, which was set to 1. Statistical significance of the differences between B73 and rta9‐1 was determined by Student's t‐tests.

2.2. Cloning RTA9 using the Mutmap Approach

To clone RTA9, we backcrossed the aphid‐susceptible mutant rta9‐1 with the wild‐type B73 to obtain rta9‐1/RTA9 F₁ seeds. The F₁ plants exhibited the same levels of aphid resistance as B73, with similar aphid survival rates and honeydew secretion levels, but significantly lower levels than rta9‐1 (Figure S2A–C, Supporting Information). Therefore, the aphid‐susceptibility phenotype of rta9‐1 may be caused by a recessive mutation.

We followed the MutMap approach^{[ 40 ]} to identify RTA9. To this end, we selected 25 aphid‐susceptible seedlings from the segregating F₂ population derived from the rta9‐1/RTA9 F₁ plants, extracted their genomic DNA, and pooled equal amounts of the DNA for library construction and genome sequencing. We identified 17809 single‐nucleotide polymorphisms (SNPs) in the bulk sequencing data, and SNP‐index analysis revealed a major peak containing 384 closely linked SNPs on chromosome 3 (Figure 2A). Of these SNPs in the candidate region, most were located in the intergenic (299), upstream (34), downstream (27), and intronic (13) regions of genes. Within the exon regions, a C‐to‐T substitution caught our attention at position 4498925, resulting in the conversion of Gln (CAG) to a stop codon (TAG). This change is predicted to lead to the premature termination of translation of the protein encoded by Zm00001d039444 (Figure 2B). We designed a derived cleaved amplified polymorphic sequence (dCAPS) marker and sequenced PCR products spanning this substitution using Sanger sequencing to assess mixed DNA pools of aphid‐susceptible and aphid‐resistant individuals from another F₂ population. The pool with the mutant phenotype was homozygous for this C‐to‐T SNP, while the other DNA pools exhibited either one wild‐type sequencing peak or a mix of two peaks at this position (Figure S3, Supporting Information).

Figure 2.

MutMap‐based cloning of RTA9 and aphid resistance tests of the rta9‐2 mutant and RTA9‐overexpressing plants. A) Distribution of CG>TA‐type mutations in rta9‐1 across the 10 maize chromosomes, shown as the SNP‐index. The red arrowhead marks the peak over the significance threshold on chromosome 3. B) Diagram of the Zm00001d039444 (RTA9) locus and confirmation of the C‐to‐T mutation in rta9‐1 by Sanger sequencing. The red box shows the codon change from CAG to TAG, a stop codon that is predicted to result in a truncated RTA9 protein. C) Survival rate of aphids feeding on B73 or the independent mutant allele rta9‐2 (^*** p < 0.001). D,E) Honeydew secretion assay. Quantification of the honeydew areas on the filter paper in (E) is presented in (D). F) Relative RTA9 transcript levels in the wild‐type KN5585 and RTA9‐overexpressing (RTA9‐OE) transgenic lines. GAPDH was used as an internal reference for the RT‐qPCR analysis. G) Immunoblot analysis indicating that RTA9‐GFP accumulates in the RTA9‐OE transgenic plants. GAPDH was used as a loading control. H) Colonization and distribution of aphids on the wild‐type KN5585 and RTA9‐OE lines after 21 days of infestation. Scale bar, 1 cm. I,J) Honeydew secretion assay. Quantification of the honeydew areas on the filter paper in (I) is shown in (J). K) Survival rate of aphids feeding on KN5585 or RTA9‐OE lines (^** p < 0.01, ^*** p < 0.001). L,M) Number and body weight of colonizing aphids after 21 days of infestation. In (F), data are shown as means ± SD, n = 3. In (C, D, J–L), the data are shown as means ± SD, n = 6. In (M), the boxes represent the interquartile range, with the middle line defining the median. The lines extending from the quartiles of the box are called “whiskers” and show the maximum and minimum values. Statistical significance was determined using Student's t‐tests (C, D, F, K) or one‐way ANOVA followed by Tukey's test (J–M) (p < 0.05), with different letters indicating significant difference.

To test whether the putative C‐to‐T mutation in Zm00001d039444 was responsible for the aphid susceptibility of rta9‐1, we ordered a Mu‐insertion mutant designed rta9‐2 from the ChinaMu database^{[ 41 ]} (Figure S4, Supporting Information). Indeed, this Mu‐insertion mutant exhibited a similar aphid susceptibility phenotype as the EMS mutant rta9‐1 (Figure 2C–E). Therefore, Zm00001d039444 appears to be the causal gene for the aphid‐susceptible phenotype observed in rta9 mutants. We hereafter refer to Zm00001d039444 as RTA9.

RTA9 is a single‐exon gene that encodes a plant‐specific Domain of Unknown Function 641 (DUF641) protein and has not been previously reported to be associated with aphid resistance. We detected transcripts for RTA9 in various B73 tissues. Relatively high expression levels were observed in the tassel, followed by the ear, with the lowest levels in the leaf (Figure S5A, Supporting Information). In addition, we examined RTA9 transcript levels at different time points following aphid infestation. The results showed that RTA9 expression initially exhibited a slight increase, followed by a transient decrease, and then peaked at 24 h post‐infestation (Figure S5B, Supporting Information).

2.3. Overexpression of RTA9 in Maize Confers Aphid Resistance without Negatively Affecting Grain Yield

To validate the role of RTA9 in aphid resistance, we generated transgenic lines that overexpress RTA9 (RTA9‐OE) in the KN5585 background, with the sequence encoding green fluorescent protein (GFP) cloned in‐frame and downstream of RTA9. We validated the transgenic lines using reverse‐transcription quantitative PCR (RT‐qPCR), which revealed that RTA9 transcript levels were significantly upregulated in the RTA9‐OE lines relative to the wild type (Figure 2F). In addition, an immunoblot analysis using anti‐GFP antibodies indicated that RTA9‐GFP accumulates in these transgenic plants (Figure 2G).

To evaluate the effect of RTA9 overexpression on aphids, we counted the number of aphids on transgenic plants. The RTA9‐OE lines had significantly fewer aphids than the wild‐type seedlings after 21 days of aphid infestation. Consistent with this result, the amount of excreted honeydew was much lower on the RTA9‐OE seedlings than on the wild‐type seedlings (Figure 2H–K). The RTA9‐OE plants had only ≈ 25% the number of aphids present on the wild type, and the aphids colonizing the RTA9‐OE plants had a lower average weight than those colonizing the wild type (Figure 2L,M). In many cases, resistance genes that enhance plant defense come with trade‐off effects.^{[ 42 ]} Therefore, we examined the agronomic and yield‐related traits of RTA9‐OE lines and the wild‐type KN5585. RTA9‐OE plants exhibited reduced plant height and ear height compared to KN5585, but had longer and wider ear leaves (Figure S6A, Supporting Information). Importantly, overexpression of RTA9 did not negatively impact seed production, including traits such as ear weight, hundred‐kernel weight, and grain weight per ear, under field conditions. (Figure S6B,C, Supporting Information). Taken together, we conclude that RTA9 overexpression improves aphid resistance in maize without negatively affecting grain yield.

2.4. RTA9 Physically Interacts with S40

To dissect the mechanism of RTA9 function, we analyzed the protein–protein interaction network of RTA9 using the STRING online tool (https://cn.string‐db.org/cgi/input?sessionId = bgtEzKim8hnP&input_page_show_search = on). We identified the senescence‐regulating protein S40, encoded by Zm00001d003009, as a potential interaction partner of RTA9.

To test the relationship between RTA9 and S40, we assessed their subcellular localization by transfecting maize protoplasts with a construct encoding RTA9 fused to GFP (RTA9‐GFP) or S40 fused to the red fluorescent protein mCherry (S40‐mCherry). We detected a dotted cellular distribution of RTA9‐GFP in cells that overlapped with the mitochondrial marker dye Mito Tracker Red, indicating that RTA9 localizes to mitochondria (Figure 3A; Figure S7, Supporting Information). By contrast, S40 was distributed throughout the cell, and was present in both the nucleus and the cytosol. The RTA9‐GFP signal partially overlapped with that of S40‐mCherry (Figure 3A), suggesting that RTA9 and S40 are partially co‐localized in mitochondria.

Figure 3.

RTA9 interacts with S40 in vivo and in vitro. A) Co‐localization of RTA9 and S40 in maize protoplasts. Mitochondria were visualized by staining with MitoTracker Red dye. Nuclei were visualized with the nuclear marker NLS‐mCherry. Free GFP was used as a control for RTA9‐GFP. B) Yeast two‐hybrid assay indicating that RTA9 interacts with S40 in yeast cells. Positive colonies were spotted onto synthetic defined (SD) medium lacking Trp and Leu (SD/−Trp−Leu) and SD medium lacking Trp, Leu, His, and Ade (SD/−Trp−Leu−His−Ade). The T + P53 combination was used as a positive control. C) Firefly luciferase (LUC) complementation imaging assay indicating that RTA9 interacts with S40 in the leaves of N. benthamiana plants. D) Bimolecular fluorescence complementation (BiFC) assay showing that RTA9 interacts with S40 in maize protoplasts. E) Co‐immunoprecipitation (Co‐IP) assay demonstrating the interaction between RTA9‐GFP and S40‐FLAG in vivo. The RTA9‐GFP transgenic line was crossed with the S40‐FLAG transgenic line, and the derived F₁ generation plants were used for Co‐IP analysis. F) His protein pull‐down assay revealing that RTA9‐MBP‐His interacts directly with S40‐GST in vitro.

We conducted a targeted yeast two‐hybrid (Y2H) assay to validate the interaction between RTA9 and S40. Yeast cells harboring pGADT7‐S40 and pGBKT7‐RTA9 were able to grow on synthetic defined (SD) selective medium (lacking Trp, Leu, His, and Ade), while cells containing all other plasmid combinations did not. This indicates that these two proteins can interact in yeast (Figure 3B). To establish whether this interaction also occurs in plant cells, we used a firefly luciferase (LUC) complementation imaging assay in the leaves of N. benthamiana plants. Leaves co‐infiltrated with nLUC‐RTA9 (encoding the N‐terminal half of LUC fused to RTA9) and cLUC‐S40 (encoding the C‐terminal half of LUC fused to S40) exhibited strong LUC activity, whereas the other combinations of constructs did not display any LUC signal (Figure 3C). A bimolecular fluorescence complementation (BiFC) assay confirmed that RTA9‐cYFP (a fusion between RTA9 and the N‐terminal half of yellow fluorescent protein [YFP]) interacts with S40‐nYFP (a fusion between S40 and the C‐terminal half of YFP) when their encoding plasmids are co‐transfected into maize protoplasts (Figure 3D). We also conducted in vivo protein co‐immunoprecipitation (Co‐IP) assays using F₁ plants harboring the RTA9‐GFP and S40‐FLAG transgenes. Following immunoprecipitation with an anti‐GFP antibody, we detected S40‐FLAG among the precipitated proteins, confirming that RTA9‐GFP interacts with S40‐FLAG in plant cells (Figure 3E). Finally, a pull‐down assay confirmed that RTA9 and S40 interact directly in vitro, as RTA9‐MBP‐His (a fusion of RTA9 and the maltose‐binding protein [MBP]) was pulled down by S40‐GST (a fusion of S40 and glutathione S‐transferase [GST]) (Figure 3F). Together, these results indicate that RTA9 can physically interact with S40 in vitro and in vivo.

2.5. Knockout and Overexpression of S40 Confirm Its Role in Maize Resistance to Aphids

S40 is a single‐exon gene. RT‐qPCR analysis detected S40 transcripts in various maize tissues, with a relatively high expression level in roots and tassels (Figure S8A, Supporting Information). Upon aphid infestation, S40 expression exhibited an early decline, peaked at 6 h post‐infestation, and then progressively decreased, reaching its lowest level at 24 h (Figure S8B, Supporting Information). To investigate the possible anti‐aphid function of S40, we generated knockout mutants of S40 in the B104 background using clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR‐associated nuclease 9 (Cas9)‐mediated gene editing. We verified the identity of one homozygous s40 knockout mutant by PCR and Sanger sequencing. The s40 mutant carries a 1‐bp insertion in the S40 coding region leading to a frameshift in the open reading frame (Figure 4A). Aphid bioassays showed that the loss of S40 function significantly enhances maize resistance to aphids; indeed, the number of aphids, the amount of honeydew secretion, survival rates, and body weight were significantly lower for aphids feeding on s40 seedlings than for those feeding on the wild‐type B104 (Figure 4B–G). Similar to the RTA9‐OE lines, we evaluated the agronomic and yield‐related traits of the s40 mutant. The mutant plants showed no significant adverse effects on key agronomic traits, including plant architecture (plant height and ear height), leaf morphology (ear leaf length and width), and seed production (ear length, ear width, ear weight, hundred‐kernel weight, and grain weight per ear) (Figure S9A–C, Supporting Information).

Figure 4.

Loss of S40 function confers aphid resistance, while S40 overexpression increases aphid susceptibility in maize. A) Diagram of the Zm00001d003009 (S40) locus and identification of the mutation in s40 by Sanger sequencing. B) Colonization and distribution of aphids after 21 days of infestation of the wild‐type B104 and s40 plants. Scale bar, 1 cm. C,D) Honeydew secretion assay of aphid feeding. Quantification of the honeydew areas on the filter paper in (C) is shown in (D). E) Survival rate of aphids feeding on the wild‐type B104 or s40 mutant (^* p < 0.05, ^** p < 0.01). F) Number of aphids present after 21 days of infestation. (G) Body weight of aphids. H) Relative S40 transcript levels in B104 and S40‐overexpression (S40‐OE) transgenic plants. GAPDH was used as an internal reference. I) Immunoblot analysis indicating that S40‐FLAG accumulates in S40‐OE transgenic plants. GAPDH was used as a loading control. J) Colonization and distribution of aphids after 21 days of infestation. Scale bar, 1 cm. K,L) Honeydew secretion assay of aphid feeding. Quantification of the honeydew areas on the filter paper in (K) is shown in (L). M) Survival rate of aphids (^*** p < 0.001). N,O) Number (N) and body weight (O) of colonizing aphids after 21 days of infestation. In (H), data are shown as means ± SD, n = 3. In (D–F, I–N), data are shown as means ± SD, n = 6. In (G, O), the boxes represent the interquartile range, with the middle line defining the median. The lines extending from the quartiles of the box are called “whiskers” and show the maximum and minimum values. Statistical significance was determined using Student's t‐test (D–H) or one‐way ANOVA followed by Tukey's test (L–O) (p < 0.05), with different letters indicating significant differences.

We generated transgenic lines that overexpress S40‐FLAG (S40‐OE) driven by the Ubiquitin promoter in the B104 background and confirmed the presence and expression of the transgenes by PCR and RT‐qPCR, respectively. The transgenic plants had a significantly higher S40 expression level than the wild type (Figure 4H) and accumulated S40‐FLAG, as shown by immunoblot analysis with anti‐FLAG antibodies (Figure 4I). The aphid resistance of S40‐OE plants was lower than that of the wild type, as the S40‐OE lines had significantly more aphids than the wild type after the initial infestation (Figure 4J). Moreover, the amount of honeydew secreted, the survival rate, and the body weight of aphids were significantly higher when the aphids fed on the leaves of S40‐OE plants than on those of the wild type (Figure 4K–O). These results confirm that S40 acts as a negative regulator of aphid resistance in maize.

2.6. RTA9 Negatively Modulates S40 Protein Stability

To elucidate the effect of the RTA9–S40 interaction, we compared the fluorescence intensity of S40‐GFP in protoplasts prepared from the wild‐type lines B73 and KN5585, as well as from the rta9‐1, and RTA9‐OE plants. The S40‐GFP fluorescence intensity was significantly higher in protoplasts from the rta9‐1 mutant than from the wild‐type B73 (Figure 5A,B). However, it was significantly lower in RTA9‐OE protoplasts than in those from the control KN5585 (Figure 5C,D), suggesting that RTA9 negatively influences the accumulation of S40.

Figure 5.

RTA9 negatively modulates S40 protein stability in vivo and in vitro. A,B) Fluorescence intensity of S40‐GFP in B73 and rta9‐1 protoplasts (A) and quantification of fluorescence levels (B). C,D) Fluorescence intensity of S40‐GFP in KN5585 and RTA9‐OE protoplasts (C) and quantification of fluorescence levels (D). E,F) Cell‐free degradation assay showing that the degradation of S40 is slower in extracts from the wild‐type KN5585 than from those of RTA9‐OE. Representative immunoblots are presented in (E) and the relative protein abundance is shown in (F). S40‐GST was detected using anti‐GST antibodies, and Coomassie Brilliant Blue (CBB) staining was used to ensure equal loading. In (B, D), the boxes represent the interquartile range, with the middle line defining the median. The lines extending from the quartiles of the box are called “whiskers” and show the maximum and minimum values. In (F), data are shown as means ± SD, and significant differences were determined with Student's t‐tests. Experiments in (E, F) were performed at least three times with similar results.

To further test whether RTA9 affects S40 stability, we performed a cell‐free protein degradation assay using recombinant purified S40‐GST produced in Escherichia coli. We added S40‐GST to a protein extract prepared from either the wild‐type KN5585 or RTA9‐OE seedlings and collected aliquots over time for analysis by immunoblotting using anti‐His antibodies. We established that recombinant S40‐GST is degraded significantly faster in the protein extracts from the RTA9‐OE lines than in those from KN5585 (Figure 5E,F). These results suggest that RTA9 negatively modulates the stability of S40 protein both in vitro and in vivo.

2.7. The Loss Function of RTA9 Results in an Imbalance of ROS Homeostasis

To determine the mechanism underlying the role of RTA9 in aphid resistance, we investigated the differentially expressed genes (DEGs) in rta9‐1 and s40 relative to their respective wild‐types B73 and B104 using transcriptome deep sequencing (RNA‐seq) analysis. To investigate the role of RTA9 in the maize response to aphids, we performed pairwise comparisons within each genetic background: aphid‐infested versus noninfested B73 (Tr_B73 versus CK_B73), and aphid‐infested versus. noninfested rta9‐1 mutant (Tr_rta9 versus CK_rta9). Finally, we defined lists of genotype‐specific DEGs by examining the overlap between the DEGs identified in B73 and rta9‐1 (Figure S10A, Supporting Information). We thus identified 281 genes as common DEGs in B73 and rta9‐1, with an additional 3193 genes specifically expressed in B73, and only 521 genes uniquely induced in the rta9‐1 mutant (Figure S10A, Supporting Information). A Gene Ontology (GO) term enrichment analysis revealed that over 200 DEGs following aphid infestation of B73 are related to oxidoreductase activity, whereas the equivalent genes were not differentially expressed in rta9‐1 (Figure S10B,C, Supporting Information). This finding indicates that aphids can induce changes in the expression of genes related to oxidoreductase activity in maize in an RTA9‐dependent manner.

Additionally, 158 DEGs were shared between B104 and s40, while 759 DEGs were specific to aphid‐infested B104 and another 959 DEGs were specific to aphid‐infested s40 (Figure S10D, Supporting Information). GO term enrichment analysis revealed that the pathway related to oxidoreductase activity ranked eighth (Q‐value, 0.0014) in the wild‐type B104 after aphid infestation, while the pathway rose to first (Q‐value, 3.4328E‐09) in the more aphid‐resistant s40 (Figure S10E,F, Supporting Information), suggesting that the functional loss of S40 alters the expression of oxidoreductase activity‐related genes in response to aphids. The DEGs shared by aphid‐infested B73 and aphid‐infested s40 (highlighted by the gray dashed line in Figure 6A) were enriched in several reported insect resistance pathways, including terpene biosynthesis, cell wall biosynthesis, JA/SA response, and oxidoreductase activity (Figure 6A,B). This result suggests that RTA9 and S40 contribute to maize aphid resistance through multiple mechanisms.

Figure 6.

RTA9 and S40 regulate the cellular redox status. A) Venn diagram showing the extent of overlap between the number of differentially expressed genes (DEGs) in the wild‐type B73, rta9‐1, B104, and s40 (control, CK; Tr, aphid infestation). The set of genes highlighted by the dashed line represents those that exhibit common changes in B73 and s40 in response to aphid infestation. B) Gene Ontology (GO) term enrichment analysis of DEGs in the region highlighted by the gray dashed line in (A). C–F) Imaging of ROS levels, detected with H₂DCFDA, in B73 and rta9 (C) and in B104 and s40 (E) under control conditions (CK) and upon aphid infestation (Tr). Quantification of the fluorescence intensity in (C) and (E) is shown in (D) and (F), respectively. G) 3,3′‐diaminobenzidine (DAB) histochemical staining of leaves from B73, rta9‐1, B104, and s40 plants under control conditions or in response to aphid infestation. H) Nitroblue tetrazolium (NBT) histochemical staining of leaves from B73, rta9‐1, B104, and s40 plants under control conditions or in response to aphid infestation. In panels (D, F), at least six individual leaves were stained and photographed. Fluorescence intensity was quantified from 100 randomly selected regions (n = 100). The boxes represent the interquartile range, with the middle line defining the median. The lines extending from the quartiles of the box are called “whiskers” and show the maximum and minimum values. Significant differences were determined using one‐way ANOVA followed by Tukey's test (D and F) (p < 0.05), with different letters indicating significant differences.

The observation that the oxidoreductase activity pathway was enriched in DEGs prompted us to investigate ROS accumulation in rta9‐1 and s40, along with their respective wild types. Accordingly, we stained the leaves with the ROS probe H₂DCFDA and quantified the resulting fluorescence intensity of its oxidized products. We determined that noninfested rta9‐1 plants accumulate significantly higher levels of ROS than the wild type B73 (Figure 6C,D). However, following aphid infestation, ROS levels rose significantly in B73 beyond those seen in rta9‐1. Aphid infestation also induced the accumulation of ROS in rta9‐1 to levels above those measured in noninfested plants, but to a lesser extent than observed in B73, resulting in significantly lower ROS accumulation than in aphid‐infested B73 (Figure 6C,D). This result was also supported by 3,3′‐diaminobenzidine (DAB) and nitroblue tetrazolium (NBT) staining (Figure 6G,H). These findings suggest that aphid infestation alters the oxidative balance in maize cells, which is dependent on RTA9. By contrast, ROS content exhibited an opposite pattern in s40, showing significantly higher levels in s40 compared to B104 regardless of aphid infestation (Figure 6E–K). We also examined ROS levels in RTA9‐OE and S40‐OE plants, as well as their respective wild‐types KN5585 and B104. Based on fluorescence intensity, the ROS content was significantly higher in RTA9‐OE lines than in the wild type (Figure S11A,B,E,F, Supporting Information), whereas S40‐OE cells had significantly lower levels of ROS than the wild type (Figure S11C–F, Supporting Information). These results suggest that the RTA9–S40 interaction may regulate aphid resistance by altering ROS homeostasis in maize.

3. Discussion

In the ongoing evolutionary arms race between plants and their insect pests, plants have evolved a diverse array of defense strategies^{[ 4} , ^{43 ]} that include the precise regulation of anti‐insect gene expression to protect against infestation, thereby enhancing their adaptive and resistance capabilities.^{[ 1} , ^{44 ]} However, the molecular mechanisms governing a substantial portion of these defense‐related genes remain poorly understood. In this study, we demonstrate that the mitochondrial protein RTA9 and its interaction partner S40 modulate ROS homeostasis to enhance resistance to aphids in maize.

During infestation, aphids first use their stylets to probe the plant epidermis before penetrating the phloem and subsequently secreting saliva into the mesophyll cells of the host. This action, along with stimuli such as physical damage and effector protein injection, triggers the production of various types of ROS.^{[ 22} , ⁴⁵ , ^{46 ]} Here, our transcriptome analysis showed that a substantial number of differentially expressed genes following aphid feeding were enriched in genes related to oxidoreductase activity in B73, but not in rta9‐1 (Figure S10A–C, Supporting Information). In agreement with this observation, ROS levels were elevated in the leaves of B73 and rta9‐1 plants following aphid feeding relative to their non‐infested controls, although they reached significantly lower levels in infested rta9‐1 than in infested wild‐type plants (Figure 6C,D,G,H). The RTA9‐OE lines accumulated higher ROS levels than their wild‐type KN5585 (Figure S11, Supporting Information), which aligns with the positive regulatory role of RTA9 in aphid resistance (Figures 1 and 2). Previous studies showed that the production ROS (H₂O₂) in plants negatively affected aphid growth;^{[ 22} , ²³ , ^{47 ]} thus, the lower ROS levels in rta9 plants may explain why aphids prefer to colonize these plants over the wild type. Additionally, we noticed that pathways related to terpene biosynthesis, cell wall biosynthesis, response to JA, and response to SA were enriched among the DEGs following aphid infestation. This observation suggests that RTA9 and S40 may contribute to the resistance of maize plants to aphids in multiple ways.

Mitochondria are semiautonomous organelles with various ETCs and are a major source of ROS in cells.^{[ 48 ]} Changes in mitochondrion‐localized proteins tend to alter the balance of intracellular ROS, which in turn affects plant resistance.^{[ 1} , ²¹ , ^{49 ]} For instance, the mitochondrial protein RTP7 modulates ROS levels by affecting the subunits of mitochondrial respiratory chain complex I, whereas the loss of RTP7 function enhances plant resistance to a broad spectrum of plant pathogens and salinity stress.^{[ 34 ]} Similarly, a mutant lacking the mitochondrion‐localized protein NATURAL BLIGHT LEAF 3 in rice accumulates H₂O₂ and exhibits enhanced resistance to biotic and abiotic stresses.^{[ 49 ]} Considering the subcellular localization of RTA9 (Figure 3A), we hypothesize that it may be involved in mitochondrial functions, such as the respiratory chain. The loss of RTA9 function likely disrupts mitochondrial electron transport, resulting in higher baseline ROS levels in the rta9‐1 mutant compared to the wild type (Figure 6C,D,G,H). Consequently, rta9 mutants accumulated ROS at a constitutively higher baseline level than the wild type (Figure 6C–H). However, upon aphid infestation, the rta9‐1 mutant showed a more modest increase in ROS levels compared to the wild‐type control (Figure 6C–H). Conversely, RTA9‐OE lines displayed elevated ROS levels and enhanced aphid resistance (Figure 2H–M; Figure S11, Supporting Information). These results demonstrate that the mitochondrial protein RTA9 is required for ROS accumulation during aphid infestation. Thus, impaired RTA9‐mediated ROS accumulation in the rta9 mutants limits oxidative bursts, ultimately leading to increased susceptibility to aphids.

Protein subcellular localization, LUC complementation imaging assays, BiFC assays, Co‐IP assays, and pull‐down analysis all indicated that RTA9 and S40 partially co‐localize and physically interact in vitro and in vivo (Figure 3). Furthermore, RTA9 negatively regulates the stability of S40 (Figure 5), which explains why these proteins affect aphid resistance in contrasting ways. For instance, rta9 mutants are more susceptible to aphid infestation than the wild‐type B73 (Figure 1), whereas the s40 mutant is more resistant than its wild‐type B104 (Figure 4). S40 is a homolog of the DUF584 family member HvS40 in barley (Hordeum vulgare), which was first reported to be associated with leaf senescence.^{[ 50 ]} HvS40 expression is upregulated under various senescence conditions, from dark‐induced senescence to age‐dependent senescence in barley leaves, making it a useful marker gene for senescence.^{[ 51 ]} To our knowledge, DUF584 has not been reported to be associated with insect resistance in plants. Here, we report that the overexpression or knockout of S40 can result in variations in ROS levels (Figure 6E–H; Figure S11, Supporting Information), which are associated with changes in aphid resistance (Figure 4). Previous studies have shown that cytoplasmic signaling regulates nuclear gene expression through retrograde signaling.^{[ 33 ]} RTA9 is localized to mitochondria and regulates S40 stability, while S40 is evenly distributed within plant cells, including the mitochondria and the nucleus, and its loss of function alters ROS homeostasis. It is therefore possible that the regulation of the RTA9–S40 module may also be achieved through retrograde signaling, a possibility that merits further investigation. The mechanisms by which RTA9 influences S40 stability should also be further explored.

Based on the above results, we propose that RTA9 mediates aphid‐triggered signaling in maize, responding to physical damage or the saliva proteins of aphids, thereby activating the signaling pathway dependent on a ROS burst. In summary, the mitochondrion‐localized RTA9–S40 protein complex is involved in regulating aphid resistance in maize, likely by modulating the plant's redox status. RTA9 overexpression and knockout of S40 confer aphid resistance but do not reduce maize seed production in the field, suggesting that RTA9 and S40 are potential candidate genes for breeding aphid‐resistant maize varieties.

4. Experimental Section

Plant Growth and Aphid Culture Conditions

Maize (Zea mays) and Nicotiana benthamiana plants were grown in a greenhouse under a 16‐h light/8‐h dark photoperiod and 28 °C/22 °C (day/night) cycles. Aphids (Rhopalosiphum maidis) were collected from maize fields in Hefei, China, and artificially bred in greenhouses for over 10 generations before use. All seedlings used for the aphid assays were cultured under a 59 × 61 × 91 cm nylon net cage to ensure there was no disturbance from other pests.

An M₂ population from an EMS‐mutagenized library in the B73 background was planted in Hainan (18°21′N, 109°10′E), China.^{[ 3} , ^{40 ]} Aphid‐susceptible and aphid‐resistant plants were scored, self‐pollinated, and harvested for study. Among them, the aphid‐susceptible mutant rta9‐1 was identified. The Mu‐insertion mutant rta9‐2 was ordered from the ChinaMu database (http://chinamu.jaas.ac.cn/Default.aspx).^{[ 41 ]} The single guide RNA (sgRNA) used for CRISPR/Cas9‐mediated gene editing was designed using CRISPR‐GE (http://skl.scau.edu.cn/home/),^{[ 52 ]} and transgenic maize lines were generated by Beijing BomeiXingao Technology (Beijing, China). Mutant plants were selected for backcrossing to the wild‐type B73, and were self‐crossed again to eliminate any other potential second‐site mutations. Homozygous mutant seeds were collected and used for the experiments.

Gene Cloning using the MutMap Method

The MutMap method^{[ 40} , ^{53 ]} was used to clone the candidate gene underlying the aphid‐susceptible phenotype of rta9‐1. The aphid‐sensitive mutant was backcrossed to the parental inbred line B73, and the resulting F₁ plants were self‐crossed. The F₂ progeny was analyzed for aphid accumulation and aphid‐susceptible individuals were identified. The leaves of 25 F₂ seedlings exhibiting the aphid‐susceptible phenotype were collected. Genomic DNA was extracted from individual seedlings using the cetyltrimethylammonium bromide (CTAB) method, its concentration quantified, and mixed in an equimolar ratio for each sample to create one pool. The resulting DNA pool was subjected to genome resequencing by Biomarker Technologies (Beijing, China). After the candidate gene was cloned and the target mutation sites were revealed, the derived cleaved amplified polymorphic sequence (dCAPS) method was used to validate the mutation. The primer sequences are listed in Table S1 (Supporting Information).

RNA Extraction, cDNA Synthesis, and RT‐qPCR

Total RNA was extracted from samples using Trizol reagent (Takara Bio, Dalian, China) and the first‐strand cDNA was synthesized with a PrimeScript RT Reagent Kit (Takara Bio) from 1 µg of total RNA following the manufacturer's protocol. For tissue‐specific expression analysis, silk and tassel samples were collected at the flowering stage, while all other tissues were collected from maize plants at the four‐leaf stage. Gene expression analysis in transgenic maize was conducted using the fourth leaf harvested from seedlings at the four‐leaf stage. qPCR was performed on a Roche LC480 LightCycler (Roche, Basel, Switzerland) using gene‐specific primers. The relative expression level of each gene was normalized to that of GAPDH by the 2^−ΔΔ Ct method.^{[ 54 ]} Each experiment was conducted with three biological replicates and two technical replicates. The sequences of all primers are included in Table S1 (Supporting Information).

Aphid Survival Rate and Honeydew Excretion

For the test of aphid survival rates, previously published methods were followed.^{[ 3 ]} To ensure that no aphids reproduced during the experiment, 20 one‐day‐old aphids were gently placed onto the second leaf of seedlings at the four‐leaf stage. The number of aphids on each seedling was determined over seven consecutive days, and the aphids were also weighed on the seventh day. Five aphids were weighed together to calculate the weight per aphid.

The amount of aphid honeydew excretion is a direct indicator of aphid feeding. For the honeydew excretion test, seedlings were cultivated with the same number of aphids as described above. After two days, a piece of perforated filter paper was placed horizontally on the seedling stalks. Two days later, the filter papers were collected, sprayed with 0.1% (w/v) ninhydrin acetone solution, and dried in an oven at 60 °C for 30 min. Finally, the color‐developed filter papers were scanned and the spot areas were quantified using ImageJ software (v. 1.54f, National Institutes of Health, Bethesda, MD, USA). Each biological experiment was independently repeated at least three times, with six leaves included per replicate.

Subcellular Localization and Co‐Localization Analysis

To investigate the subcellular localization of RTA9 and S40, their full‐length coding sequences without the stop codon were cloned individually into a modified pCAMBIA1305 vector containing a green fluorescent protein (GFP) sequence and the cauliflower mosaic virus (CaMV) 35S promoter.^{[ 55 ]} The S40 coding sequence was also cloned in‐frame with the sequence of the red fluorescent protein (RFP) in pCAMBIA1300. Maize protoplasts were isolated through enzymatic digestion and the plasmids were transferred into the protoplasts using the polyethylene glycol (PEG) method.^{[ 54 ]} After incubation for 24 h in the dark, GFP and RFP fluorescence signals were captured using a laser confocal microscope (LSM800; Carl Zeiss, Oberkochen, Germany).

Yeast Two‐Hybrid Assay (Y2H)

The full‐length coding sequences of RTA9 and S40 were cloned into the pGAD‐T7 and pGBK‐T7 vectors, respectively. The resulting plasmids were introduced into the yeast (Saccharomyces cerevisiae) strain AH109. Positive transformants were selected for growth on synthetic defined (SD) medium lacking Trp and Leu at 30 °C for three days. Positive colonies were resuspended in sterile ddH₂O and washed with sterile ddH₂O at least three times, after which the cells were spotted onto SD medium lacking Leu, Trp, His, and Ade and cultured at 30 °C for three days.

Firefly LUC Complementation Imaging Assay

For the LUC complementation assays, the full‐length coding sequences of RTA9 and S40 were cloned into the appropriate cloning sites of the nLUC or cLUC vector, respectively, using restriction enzyme digestion and ligation methods.^{[ 56 ]} The resulting plasmids were introduced individually into Agrobacterium (Agrobacterium tumefaciens) strain GV3101 containing pSoup + p19. Pairs of Agrobacterium cultures harboring each plasmid were resuspended in infiltration buffer (0.01 M MES, 0.01 mM MgCl₂, 0.1 mM acetosyringone) and mixed in a 1:1 ratio before being co‐infiltrated into the leaves of N. benthamiana plants. After three days of culture in an incubator, the leaves were uniformly sprayed with 1 mM D‐luciferin, and the LUC signal was captured with a chemiluminescence camera (5200; Tanon Science and Technology, Shanghai, China).

Bimolecular Fluorescence Complementation (BiFC) Assay

For the BiFC assays, the full‐length coding sequences of RTA9 and S40 were cloned into pCAMBIA1300‐YCE or pCAMBIA1300‐YNE vectors, respectively, at appropriate cloning sites via restriction digestion. The resulting plasmids were transfected into maize protoplasts using the PEG method. The transfected cells were cultured for 24 h at 26 °C in the dark, after which YFP signals were observed by confocal fluorescence microscopy (LSM800; Carl Zeiss, Oberkochen, Germany).

Recombinant Protein Production and In Vitro Protein Pull‐Down Assays

For the in vitro protein pull‐down assays, RTA9‐MBP‐His and S40‐GST were produced in E. coli.^{[ 56 ]} Briefly, the full‐length coding sequence of RTA9 was inserted into a modified pET28a(+) vector containing the MBP sequence; the full‐length coding sequence of S40 was cloned into a pGEX‐6P‐1 vector containing the GST sequence. The resulting plasmids were introduced into E. coli BL21(DE3) and the production of each recombinant protein was induced by incubation at 16 °C for 24 h in the presence of 0.1 mM isopropyl‐β‐D‐thiogalactopyranoside (IPTG). The cell pellets were collected by centrifugation at 4 °C, and then lysed by sonication (XM‐650T, Jingxin, China). The proteins in the supernatant were collected by centrifugation and purified with Ni²⁺‐NTA and glutathione agarose beads (Solarbio, Beijing, China). RTA9‐MBP‐His or the negative control MBP‐His was mixed with S40‐GST, and the mixtures were then added to 5 µL of His trap beads (ChromoTek, Planegg, Germany) and incubated at 4 °C for at least 4 h. The input and pulled‐down proteins were detected using anti‐His (Roche) or anti‐GST antibodies (EnoGene), respectively.

In Vivo Co‐Immunoprecipitation (Co‐IP) Assays

For in vivo Co‐IP assays, the RTA9‐GFP transgenic line was crossed to the S40‐FLAG transgenic line to obtain F₁ RTA9‐GFP S40‐FLAG plants. Then, 25 g of these F₁ seedlings at the four‐leaf stage were collected and ground into fine powder in liquid nitrogen and then resuspended in 25 mL of extraction buffer (50 mm Tris‐HCl, pH 8.0; 150 mm NaCl; 0.1% [v/v] IGEPAL; 2.5 mm EDTA, 10% [v/v] glycerol; 10 mm β‐mercaptoethanol; 1 mM PMSF; 10 µm leupeptin; and 1× Roche protease inhibitor cocktail [Roche, Basel, Switzerland]).^{[ 56 ]} The resulting mixture was incubated on ice for 10 min and then centrifuged at 12000 × g for 20 min at 4 °C. The supernatant was further clarified by repeating the centrifugation until it was clear. GFP Trap beads, pre‐washed with extraction buffer, were then added to the protein extracts and incubated at 4 °C for 3 h. Following three washes with extraction buffer, the beads were collected, boiled in SDS loading buffer, and subjected to SDS‐PAGE for immunoblot analysis. Target proteins were detected using anti‐GFP (Roche, Basel, Switzerland) or anti‐FLAG (MBL, Sapporo, Hokkaido, Japan) antibodies.

In Vitro Protein Degradation Assays

Recombinant purified S40‐GST was incubated with total protein extracts from B73 or rta9‐1 to assess its stability. Briefly, maize leaves were ground to a powder in liquid nitrogen and their total proteins were extracted by adding 1 mL protein extraction buffer (50 mm Tris‐HCl, pH 7.5, 150 mm NaCl, 0.1% [v/v] IGEPAL, 2.5 mm EDTA, 10% [v/v] glycerol, 10 mm 2‐mercaptoethanol, 10 µm leupeptin) to each 1 g of powder. The proteins in the supernatant (soluble proteins) were obtained by centrifugation at 12000 ×g for 20 min at 4 °C, after which ≈ 20 µg of recombinant S40‐GST was added to 1 mL of maize protein extract and incubated at 28 °C for different durations. S40‐GST abundance was detected using an anti‐GST antibody (Abmart, Shanghai, China) after SDS‐PAGE and transfer to PVDF membrane.

Transcriptome Analysis

For RNA‐seq analysis, maize seedlings were grown to the four‐leaf stage in the greenhouse before 20 adult aphids were moved onto their second leaves for 24 h. The leaves and sheaths from six independent maize seedlings were collected as one experimental replicate. Total RNA was extracted from the samples using an RNeasy kit (Thermo Fisher Scientific, Waltham, MA, USA). Three independent experimental replicates were performed for each sample. The control group (CK) consisted of plants not infested with aphids, defined as CK_B73, CK_rta9, CK_B104, and CK_s40. The treatment group (Tr) comprised aphid‐infested plants, designated as Tr_B73, Tr_rta9, Tr_B104, and Tr_s40.

cDNA libraries were sequenced using a NovaSeq 6000 instrument (Illumina, San Diego, California, USA) by Lianchuan Biotechnology Company (Hangzhou, China). The sequencing and analysis methods were described previously.^{[ 10 ]} The fragments per kilobase of exon per million reads mapped (FPKM) values were calculated to normalize the expression data, and the R package DESeq2 was used to determine differential expression among samples.

Clean reads were mapped to the maize B73_v5 reference genome. Differentially expressed genes (DEGs) were defined as those with an absolute fold‐change ≥1.5 with q < 0.05. The identified DEGs were subjected to GO term enrichment analysis using the Lianchuan cloud tools (https://www.omicstudio.cn/tool).

Determination of ROS, H₂O₂, and Superoxide Anion (O₂ ⁻) Contents

To assess the effect of RTA9 and S40 on ROS, the accumulation of ROS, H₂O₂, and O₂ ⁻ was quantified following the methods described in previous studies.^{[ 57} , ^{58 ]} Briefly, leaves were soaked in staining buffer (10 mm Tris‐HCl, 50 mm KCl, 50 µm H₂DCFDA, 0.02% [v/v] Tween‐20, pH 7.2) and subjected to vacuum infiltration at −0.6 MPa for 20 min. Excess dye was removed by rinsing the leaves with distilled water, and fluorescence was captured and quantified using a laser confocal microscope (LSM800; Carl Zeiss, Oberkochen, Germany). H₂O₂ was detected using 3,3′‐diaminobenzidine staining (DAB, Sigma–Aldrich, St. Louis, MO, USA). Leaf segments were treated with DAB solution (1 mg mL⁻¹) for 10 h, followed by decolorization in 80% (v/v) ethanol, which was refreshed until it turned colorless; the samples were observed under a microscope (Leica M165, Germany). O₂ ⁻ was visualized through staining with nitroblue tetrazolium (NBT, Servicebio, Wuhan, China). The leaf pieces were incubated in 20 mL of NBT solution (50 mg dissolved in Tris‐HCl buffer pH 7.4) for 10 h, decolorized in 80% (v/v) ethanol, and then visualized under a microscope (Leica M165, Germany).

Assessment of Agronomic and Yield Traits

To comprehensively evaluate the agronomic and yield traits of the RTA9‐OE and s40, plants were grown at a density of 25 × 60 cm.^{[ 59 ]} A randomized block design was employed with two replicates per line. In each replicate, data were collected from two randomly selected rows, with at least 15 individual plants per genotype, excluding the first and last plants in each row. Agronomic and yield traits were assessed as follows: Plant height was measured as the distance from the soil surface to the tip of the tassel at anthesis. Ear height was measured as the distance from the stem base to the node bearing the uppermost ear. Ear leaf length was recorded as the distance from the tip of the ear leaf to the top of its sheath. Ear leaf width was recorded at the maximum width of the ear leaf midpoint. Ear length was measured as the total length of the mature ear, including both filled and unfilled portions. Ear diameter was measured at the widest point of the mature ear. 100‐kernel weight was determined by weighing 100 randomly selected kernels after drying. Ear weight was determined as the dry weight of the ear, including cob and kernels. Grain yield per plant was determined as the dry weight of all kernels per plant after threshing. For each trait, at least 15 independent plants were assessed, except for 100‐kernel weight, which was measured with six independent replicates.

Statistical Analysis

All data obtained in this study were analyzed using SPSS Statistics 20.0 software (IBM, Armonk, New York, USA) and GraphPad Prism 8.0.1 (San Diego, CA, USA). For aphid bioassays (population counts, body weight, survival rate, and honeydew excretion), at least six independent biological replicates were performed per experiment. Data are presented as means ± standard deviation (SD), except for body weight (box plots). For RT‐qPCR, experiments were independently performed three times with results expressed as means ± SD. For ROS detection, at least six individual leaves per treatment were stained and photographed, with fluorescence intensity quantified in 100 randomly selected regions. Statistical significance was determined using two‐tailed unpaired Student's t‐tests or one‐way analysis of variance (ANOVA) followed by Tukey's test, with p < 0.05 considered significant.

Conflict of Interest

The authors declare no conflict of interest.

Author Contributions

C.H.W. and X.Q.Z. contributed equally to this work. P.J.L. conceived and supervised the project. P.J.L. and C.H.W. designed the experiments and analyzed the data. C.H.W., X.Q.Z., L.W., S.J.H., J.M., T.Y.W., Y.B.Z., J.H.D., C.C., Z.T., X.Z.C., and N.L. performed the research. C.H.W. wrote the manuscript.

Supporting information

Supporting Information

Wang C., Zhang X., Tao Z., et al. “The Mitochondrial Protein RESISTANCE to APHIDS 9 Interacts with S40 to Resist Aphid Infestation by Modulating Reactive Oxygen Species Homeostasis in Maize (Zea mays).” Adv. Sci. 12, no. 46 (2025): e04382. 10.1002/advs.202504382

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.